TW201004370A - Image signal correcting device, imaging device, image signal correcting method, and program - Google Patents

Abstract

Provided is an image signal correcting device (15) comprising a color mixing correction unit (151) for correcting a color mixing component contained in an image signal, which has been photo-electrically converted by a color imaging element (12), in response to the image signal, and a storage unit (153) for storing a color mixing correction coefficient which has been set in advance for each of the regions divided in plurality to correspond to the imaging plane of the imaging element (12). The color mixing correction unit (151) approximates the color mixing correction coefficient, which has been read out of the storage unit (153) at the color mixing correction time, by an interpolation, to thereby acquire a color mixing percentage at a necessary pixel position, so that the color mixing component is corrected by using the color mixing percentage acquired.

Description

[Technical Field] The present invention relates to an image signal correcting device for correcting color mixing (color crosstalk) between adjacent pixels in an image pickup apparatus having a solid-state image sensor such as a CMOS image sensor, Camera, image signal correction method and program. [Prior Art] In general, an imaging element such as a CMOS (Complementary Metal Oxide Semiconductor) image sensor (CIS) generates color mixture (color crosstalk) which is affected by adjacent pixels. This color is roughly classified into an optical color mixture in which a diffusion charge is mixed in the substrate, and an optical color mixture in which incident light leaks into an adjacent pixel due to disordered reflection or the like. These color mixing is a factor that adversely affects the image f such as deterioration in color reproducibility. a. Therefore, there is a technique for making the color correction for each color, a technique for correcting the in-plane law, or a technique for correcting the same according to the F value (for example, refer to the patent document). Simultaneously with the color mixing, there is no absorption in the pixel due to the absorption of light, etc., which does not contribute to the light-ray--the amount of light that is converted first, and produces vignetting in units of pixels. It affects the interface reflection or has the same composition and color mixture as the color mixture. The in-plane part is called vignetting, such as pixel shading, light absorption of polycrystalline enamel, etc., cloth, which becomes the main cause of deterioration of enamel: people propose to have only shadows about the pixel, consider imaging Correction device for in-plane distribution (refer to the special boring tool, 3). These are systems for shading correction of the signal from the sensor output that ignores the 138321.doc 201004370 color. [Patent Document 1] 曰本特开平10 [Patent Document 2] JP-A-2003-169255 [Patent Document 3] JP-A-2006-134157 (Summary of the Invention) [Problems to be Solved by the Invention] However, In a compact camera such as a telephone, the difference in the incident angle between the central pixel of the sensor and the peripheral pixel is increased by the miniaturization of the pixel pitch and the short distance of the exit pupil distance (the more the peripheral principal ray is inclined) Therefore, the closer to the periphery, the more light is mixed in adjacent pixels, and the pixel unit from the microlens to the photodiode to the adjacent pixel A1 wiring structure is asymmetrical, so the light is reflected to arrive. Photodiodes are more complicated before.

It is difficult to correct the color mixture in the face.

It is not equivalent in mechanics.

Quantity. It is difficult to estimate the correct signal only by considering the dissipation of light. 138321.doc 201004370 Therefore, the above-mentioned yin> correction system is difficult to perform high-precision correction. The present invention provides an image signal correction splitting, an image pickup apparatus, an image signal correction method, and a program which can realize in-plane-law color mixing correction and can realize high-precision correction. [Technical means for solving the problem] The image signal correcting device according to the first aspect of the present invention has at least a color mixing correction unit that receives an image signal photoelectrically converted by a color image sensor, and corrects a color mixture component contained in the image signal. And a memory unit that stores a color correction coefficient that is set in advance in each of the plurality of regions corresponding to the imaging surface of the imaging element; and the color correction portion is approximated by the memory portion by interpolation The mixed color correction coefficient is read, the color mixing ratio of the necessary pixel position is obtained, and the mixed color ratio is used to correct the color mixture component. An imaging device according to a second aspect of the present invention includes: a color imaging element, an imaging subject image, and an image signal correcting device that receives an image signal photoelectrically converted by the color imaging element and corrects the image; and the image The signal correction device has at least: a color mixing correction portion that corrects a color mixture component contained in an image signal photoelectrically converted by a color image; and a memory portion that is stored in advance and is plurally divided on an imaging surface corresponding to the image pickup device The color mixing correction coefficient of each of the regions is further obtained by the color mixing correction coefficient read by the memory portion by the interpolation approximation when the color mixture correction is performed, and the color mixture obtained by the color mixing ratio is obtained. The rate is corrected for the color mixing component. Preferably, the color mixing correction unit sets a representative point of the plural number, and the inner color ratio is determined by inserting between 138321.doc 201004370 to represent the color mixing ratio of each pixel. The car is better. The color mixing correction unit performs the sum of the obtained color mixing rate and the input image signal, and performs the color mixing correction by subtracting the input image signal and the product and the difference result. Preferably, the color mixture correction coefficient stored in the memory portion is measured and stored in advance as a color mixture ratio of the other color pixels from the one-shirt color pixel. Preferably, the image compensation unit further receives an image signal corrected by the color correction of the color mixture correction portion, and performs pixel shading correction on the color correction image signal.乂有|||There is a second memory unit that stores a pixel shading correction coefficient that is set in advance in each of the plurality of regions corresponding to the imaging surface of the imaging unit 7L, and the shading correction portion is approximated by interpolation The pixel shading correction coefficient stored in the second memory portion is obtained, and the correction coefficient of the necessary pixel position is obtained, and the shading correction is performed by using the obtained correction key. Preferably, the shading correction unit performs shading correction by integrating the obtained correction coefficient with the mixed color correction image signal. According to a third aspect of the present invention, in the image signal correcting method, the step of reading the color mixture correction coefficient set in each of the regions corresponding to the imaging surface of the color image sensor is performed by the memory unit; The image signal of the photoelectric conversion of the component, the step of obtaining the color mixing ratio of the necessary pixel position by approximating the color mixing correction coefficient read by the memory portion by the internal broadcasting; and correcting the color mixing component by using the color mixing ratio obtained by the upper connection A step of. The program of the fourth aspect of the present invention causes the computer to perform image signal correction 138321.doc 201004370 processing, and the image signal correction processing has the following steps: the memory unit reads the imaging surface preset to the corresponding color imaging element and divides the plurality Processing of a color mixing correction coefficient of each region; receiving an image signal photoelectrically converted by the imaging element, and approximating a color mixing correction coefficient read by the memory portion to obtain a color mixing ratio of a necessary pixel position; And a process of correcting the color mixture component by using the color mixture ratio obtained as described above. According to the present invention, the image signal photoelectrically converted by the color image pickup element is rotated into the color mixture correcting portion. When the color correction is applied to the color mixture correction, the color mixture correction coefficient read by the memory unit is approximated by interpolation, and the color mixture ratio of the necessary pixel position is obtained. Further, the color mixture correcting unit complements the color mixture component by the obtained color mixture ratio. [Effect of the Invention] According to the present invention, color mixture correction can be realized based on the in-plane distribution, and the color correction can be corrected with high precision. [Embodiment] Hereinafter, embodiments of the present invention will be described with reference to the drawings. Fig. 1 is a block diagram showing an example of the configuration of an image pickup apparatus to which image signal compensation according to an embodiment of the present invention is applied. As shown in FIG. 1, the imaging device 1 includes a lens system u, an imaging element 12, a her converter 13, a clamp unit H, an image signal correcting device image 15, a demosaicing unit 16, and a linear matrix portion 17. The gamma correction unit a and the chromaticity signal generation unit 19 are included. The video interface 138321.doc 201004370 (I/F) 20 is disposed on the output side of the redundant luminance chrominance signal generating unit 19. The lens system (10) contains an aperture (1) which images the subject as an imaging surface of the image 12.摄像 The imaging element 12 is formed by a CMOS image sensor or the like, and has a plurality of unit pixels arranged in a matrix. A color filter is disposed corresponding to each pixel. Camera component! 2, as its pixel arrangement, as shown in Figure 2. Fig. 3 is a circuit diagram showing a configuration example of a unit pixel of the image sensor 12 of the embodiment. Fig. 3 is a view showing an example of a pixel of an image sensor composed of four transistors in the embodiment. Each of the pixels 120' has a photoelectric conversion element 12 1 such as a photodiode as shown in Fig. 3 . Further, the pixel 120 has four transistors of the transfer transistor 122, the reset transistor 123, the amplification transistor 124, and the selection transistor 125 as the active element for the one photoelectric conversion element 121. The photoelectric conversion element 121 photoelectrically converts incident light into an electric charge (here, electron) corresponding to the amount of light. The transfer transistor 122 is connected between the photoelectric conversion element 丨2丨 and the floating diffusion FD, and the gate (transfer gate) is supplied with the control signal transmission signal TG via the transfer control line LTx. Thereby, the electrons to be photoelectrically converted by the photoelectric conversion element m are transferred to the floating diffusion FD by the transfer transistor 122. 138321.doc 201004370 The reset transistor 123 is connected between the power supply line LVDD and the floating diffusion FD, and is supplied with a reset signal RST of the control signal via its reset control line LRST. Thereby, the reset transistor 231 resets the potential of the floating diffusion FD to the potential of the power supply line LVDD. The floating diffusion F D is connected to the idle electrode of the amplifying transistor 12 4 . The amplifying transistor 124 is connected to the signal line LSGN via the selection transistor 125 to form a constant current source 126 and a source follower outside the pixel portion. And, via the selection control line LSEL, the selection signal SEL corresponding to the address signal as the control signal is applied to the gate of the selection transistor 125, and the selection transistor 125 is turned on. When the selection transistor 125 is turned on, the amplifying transistor 124 amplifies the potential of the floating diffusion FD, and outputs a voltage corresponding to its potential to the signal line LSGN. The voltage output from each pixel is output to the line read circuit via the signal line LSGN. In these operations, since the gates of the transfer transistor 122, the reset transistor 123, and the selection transistor 125 are connected in units of columns, the pixels of one column are simultaneously performed. The image pickup device 12 is formed by arranging the reset control line LRST, the transfer control line LTx, and the selection control line LSEL wired in the pixel array unit in units of columns of the pixel array. The reset control line LRST, the transfer control line LTx, and the selection control line are driven by a vertical scanning circuit (not shown). Further, the signal line LSGN is connected to a line reading circuit including a CDS circuit (correlated double sampling power 138321.doc 201004370). Fig. 4 is a schematic view showing the structure of the device of the image pickup device and an explanatory view of the color mixture. In FIG. 4, the image pickup device 12 has a light-receiving surface for the photodiode PD of the photoelectric conversion element, and the color filter CF is disposed with the waveguide WGD interposed therebetween, and is positioned on the light incident side of each color filter CF. The multilayer wiring structure of the lens ML. An A1 wiring, an interlayer film, or the like is formed on the waveguide 160D. Color mixing is more common in 7L parts of a multi-layer wiring structure such as a CMOS image sensor. The main reason is that the components of the obliquely incident light are caused to pass through the A1 light-shielding film and leak into the adjacent pixels. This is referred to herein as optical color mixing. In addition, there is also a color mixing of the photodiode PD by electron diffusion, etc., but the main reason for non-mixing in the current process. The color mixing phenomenon is achieved by the balance of the light emitted by the peripheral pixels or toward the peripheral pixels due to the disordered reflection in the waveguide WGD forming the path color filter CF to the path of the photodiode PD to reach the photodiode A phenomenon caused by the complexity of the PD. Since the structure of the pixel is generally asymmetrical between adjacent ones, it is also difficult to estimate the disordered reflection. The color correction and shading correction of the present invention will be described in detail later. The A/D conversion β 13 ' is converted into a digital signal by the image signal of the image pickup device 12 and output to the clamp portion 丨 4 . The clamp portion 14 corrects the black level of the digital image signal by the A/D converter 13 and outputs it to the image signal correcting means 15. 138321.doc 201004370 The image signal correcting device 15 is an effective color mixing ratio θ estimated in each of the regions divided in a matrix (mesh shape) in accordance with the imaging surface of the imaging device 12, and the color mixing ratio is used therein. The θ on the necessary pixel position (coordinate) is obtained by approximating, and the color mixture correction associated with θ is performed. Image H; correction device 15 performs shading correction on the image signal for performing color correction. The image signal correcting device 15 is divided into a matrix (mesh shape) by using an imaging surface (correction target surface) that matches the imaging element 12, for example, when the color correction and the shadow correction are performed, and the number of the first unit of the mesh unit + & The representative point is interpolated with a so-called B-spline (B-SpHne) with respect to each weighting coefficient. As shown in Fig. 1, the color mixture correction unit, the first memory unit 153, and the second memory unit image signal correcting means 15, 151 and the pixel shading correction unit 152 154 are provided. . The hidden 4153 stores the effective color mixing ratio 1 estimated by each of the areas divided by the matrix (mesh shape) of the imaging surface of the imaging element 12.

The two-color complement 151 ′ is calculated by interpolating the one of the pixel positions (coordinates) necessary for the color correction stored in the color ratio 记忆 of the memory unit 153, and the color mixture correction is performed by the approximate color mixture ratio Θ. The color correction unit 151 performs interpolation of the B-spline of the color mixture ratio (correction coefficient) stored in the storage unit 153 to perform color mixture correction. (10) The 卩154 is a correction coefficient defined by each of the divisions (the mesh shape) of the imaging element η. = The positive system performs the correction of the correction coefficient stored in the memory unit 154. The spline is interpolated to perform shading correction. 138321.doc -11 - 201004370 The color correction and shading correction of the image signal correction device 15 will be described in detail later. The image signal correcting means 15 outputs the shading-corrected image signal to the demosaicing unit 16 after the color mixture correction. After the color mixture is corrected, the image signal corrected by the shading is subjected to the synchronization processing in the demosaicing unit, and the color reproduction processing is performed in the linear matrix portion n, and the gamma correction portion 18 is subjected to gamma correction, and is supplied to the luminance color. The luminance signal generating unit 19 generates a luminance signal and a chrominance signal in the luminance chrominance signal generating unit 19, and displays the video on the display device (not shown) via the video interface I/F 20. Hereinafter, a description will be given of the principle to specifically describe The image color correction device 15 combines the color correction and the shadow correction. Fig. 5 is the image of the light dissipation and absorption process. Fig. 6 is the image of the light dissipation and absorption inverse process. Dissipation absorption process of light q❶ before photoelectric conversion (Qo~>q0) Figure 6 shows the inverse process of incident light Q〇 to the pixel 〇). It is shaped, dissipated process (vignetting) and absorption process (mixed color) is not the same. That is, the dissipation process (vignetting) # absorption process (mixed color). 'Kun color, as shown in Figure 5, depends on the absorption process of light from surrounding pixels, the shadow depends on the light of the object pixel Scatter The process is not thermodynamically equivalent. Therefore, as shown in Fig. 6, when the light is dissipated in the reverse direction, the absorption process must first correct the absorption from the surrounding pixels (referred to as the color mixture 13832J.doc). -12- 201004370 Into: positive), and then the order of the amount of dissipation of the positive light (referred to as the pixel shading correction) = It can be said that the shadow correction system of the existing technology only considers the dispersion of light without considering the degree of displacement of the absorption process. Therefore, it is difficult to estimate the correct machine number. That is, 'mixing color and pixel shading are necessary to exaggerate the two to ensure that the system is essential. ^ It can be seen that the color correction and shadow correction must be independent and correct the color to correct a yin shirt correction The image signal correcting device 15 of the present embodiment is constructed by shading correction of the mixed color line. Here, the optical color mixing model is modeled. Fig. 7 is a view showing a color mixing model between pixels. A diagram of the dissipation of light and the absorption model of each pixel. To indicate the light quantity balance of the target pixel nickname of FIG. 7, the optical path: the path of the filter to the path of the photodiode) Scattered, sucked: The quantity 'is the same as the color through the color filter, the slit after the light film (4) W, the price is defined as shown in Figure 8. ^ξ, here the amount of light that is not dissipated by the pixel i to the surrounding pixels or The voltage ' ^ indicates the amount of light or voltage incident from the ritual reflection (= pixel i = light or voltage), qi indicates the sensor output signal amount of the pixel i (light * voltage or digital signal value). (A) and (B) show a diagram showing an example of the rounding of the imaging element (sensor) 12. Fig. 9 (4) shows the color crossing of the imaging element (sensor) 12 138321.doc • 13- 201004370 After passing the input signal, Figure 9(B) shows the image sensor (sensor) to output the signal. If the color mixing model is normalized as shown below, then [Model 1] (Formula 1) The model of the pixel signal amount ^ 'q. Table (4) The voltage of the county pixel after photoelectric (four) or the digital signal after AD conversion. :. : indicates the amount of digital signal after the amount of light passing through the color pupil of the target pixel is converted by voltage or AD is converted. The light is converted into a voltage amount by the color of the color directional film adjacent to the pixel of the target pixel, or is converted into a converted digital signal amount. ...the amount of light that is dissipated by the object pixel toward the periphery does not contribute to the non-human 7L coefficient of the amount of incident light of the target pixel by the amount of light of the Qg in the object image _ 汛. In general, 〇〇 <"< 〇." indicates the amount of light that is scattered from the periphery to the target pixel (= other than Q: the amount of light of the peripheral Qj of the signal amount q of the target pixel) divided by the incident The non-dimensional coefficient of the amount of light Q. Generally, 0_0<η』〇<ι·〇. The 5C〇 table does not balance with the absorption coefficient, the target pixel and the surrounding pixels _ 尤 (Example: towards the image ^ does not contribute to the signal amount of the target pixel q 吸收 the absorption rate of the absorption of the polycrystalline silicon around the pixel, etc.), here, 13832l.doc -14- 201004370 〇<x〇<l. Indicates the number of pixels adjacent to the target pixel. The expression 1 is deformed as shown in Equation 2 to indicate the amount of signal incident on the pixel. [Number 2] ~ Σ^οβ,·

M J ··_ (Formula 2) 1~t4〇J~z〇 The amount of signal of the incident light of the pixel... The deformation of Equation 1 shows that the amount of signal corrected by the scattered light in the optical system path is not absorbed. The optical path is an optical path in the waveguide WGD in which the color filter CF of Fig. 4 is formed to the path of the photodiode PD. That is, the denominator of Equation 2 represents the light dissipation, and the pixel shading correction for correcting the amount of the signal component that does not contribute to the target pixel.

The numerator of the formula ^ indicates that the light scattered by the surrounding pixels is corrected to contribute to the color mixing amount of the signal component of the object pixel. Here, the shadow of the lens is the same as the pixel shadow of the so-called pixel. The second is to describe the formalization of the mixed color. The signal quantity Cg after the color correction is modified by the following formula [3] ^ 7=1

···(Expression 3) 138321.doc 201004370 The second term on the right side of Equation 3 refers to the sum of the neighboring pixels (n) of the target pixel No. 0 and SQ. According to the model of the present invention, if the color mixture correction is performed first and the pixel shading is performed first, the Q 对象 of the target pixel is obtained. However, since Qj on the right side of Equation 3 is the amount of light passing through the color light-emitting sheet shown in Fig. 9(A), it is generally an unmeasurable amount. Therefore, it is necessary to separately expand the measurable sensor output shown in Fig. 9 (8) as shown in the following formulas 4 to 6. Fig. 1 is a flow chart showing the signal processing of the image signal correcting device 15 of the present embodiment. Rational picture. As shown in Fig. 10, 'the image of the rear stage can be correctly displayed. The image signal correction device i 5 of the present embodiment is implemented by performing a color correction pixel on a sensor output q〇. Shadow correction. If this signal processing (skip) is not performed, άw A ..., , , / ) is known from Equation 3, and the Σηρ term enlarges the degree of shadow enhancement and Q〇 eventually cannot be reproduced correctly. From the above point, when Equation 2 is developed by the adjacent pixel measurable quantity q, Equation 4 below is obtained. ‘' [number 4]

_ ·_·(7) 4) As shown in Fig. U, it is implemented in the order of the far side of the target pixel. 138321.doc 201004370 Adjacent, thin, human post, next adjacent image positive, most connected, at ~ *1冢f When the color correction and the shading complement the formula 3 of the gradation of the correction of the target pixel nickname, the RGB measurable amount qj is developed using the measurable amount qj of the adjacent pixel. ···(式5) Σ^〇7·~^〇)0 where

Ac ^j〇

~^jk ~XJ In addition, the day w w will be the completion of the 5th color to the adjacent and sub-adjacent pixels, the following formula. [Number 6] H 9. Let + be β be household 1 Μ *=1 η η η St. Buy βΛ&|...(Formula 6) >»Ι Ar«i /*1 ΣΣΣβ/〇β^β/*ί/+ I can not be determined The quantity Q to the right side of the right side of the mysterious form 6 contains the unmeasurable quantity Q. Further, when the dispersion balance of each pixel is at most the number of light incident on the pixel, the value of η + Σξ + Χ < 1 holds. <<1' Mixed color. The less the shadow of the pixel, the sensor (imaging by β < 1 ' is known from Equation 6] The correction c can be expressed sufficiently by the approximate value of q-XPq. It can be said that the more the β element is, the better. Therefore, the basic color mixing in the embodiment of the present invention is officially the underlined portion of Equation 7 138321.doc 17-201004370. [Number 7] η ^ Σ ^>Oj - %〇)00 = - 5] Py〇?y Μ (Expression 7) The color mixture is added, and the addition part of the second term on the right side of Equation 7 (Σ|3〇 is the object) Therefore, for example, the imaging element 12' in which the color filters of red (R), green (G), and blue (Β) are arranged in a mosaic can be theoretically obtained by the following method. After the color mixing amount, ρ - θ is defined, and the addition portion of the second term on the right side of Equation 7 is rewritten to 2 eq. Thereafter, 0 is redefined as the effective color mixing ratio. <Color mixing rate 求 method > : Fig. 12 shows Fig. 1 is a diagram showing the response and color mixing amount of each pixel of the monochromatic light r. The configuration of the color mixing amount measuring system 30 of Fig. 12 includes a white light source 31 and a belt. The wave-passing device (BPF) 32, the lens 33, and the image sensor 34. The image sensor 34 has the same configuration and function as the image sensor 12 of Fig. 1. For example, as shown in Fig. 12, the sensitivity of each color filter CF is prepared. Monochromatic light of the band pass filter 32 near the maximum wavelength. Monochromatic light of each color is irradiated to the image pickup element 34 via the lens 33, The response amount of each pixel is obtained. Then, the color mixture amount is obtained from the pixel values obtained by the respective color light sources. That is, as in the example of Fig. 13, when the monochromatic light scale is irradiated, the band pass filter is 138321.doc 18 201004370

The feature does not contribute to the case of G pixels and B pixels. Since substantially G pixels and B pixels are absorbed by the g and b shirts (10), they do not directly respond to & Further, the early-color light R (the light source having the maximum sensitivity of the color light-receiving sheet R and having a limited wavelength band) is realized by passing the white light source through a band pass filter near the R wavelength or the like. In this case, as shown in Fig. 13, the response amount (signal amount) existing in the 0 pixel and the B pixel corresponds to the color mixture amount from the adjacent R pixel. Similarly, G light and B light are also applied to each of the G pixel and the b pixel, whereby the amount of color mixture can be measured. For example, the color mixing ratio from the R pixel to the G pixel can be obtained by the following Equation 8. [Equation 8] The color mixture ratio from the R pixel to the G pixel is 0rg = a value of G at the time of monochromatic light irradiation (Equation 8) The value of the pixel R at the time of R monochromatic light irradiation ^ Similarly, G can be obtained Pixels to R pixels, G pixels to B pixels, ... and so on. V, and the color mixing of the local target form is formal, and can be expressed by the following formulas 9 and 10. [Number 9] Mixed color complement formal c〇=%~ieJ〇qj ...u 9) [Number 10] The value of pixel 0 is the value of pixel j. The denominator is the source of pixel j only through the color light sheet. The value at the time of irradiation. ···(式10) 138321.doc -19- 201004370 So far, the memory part method. The amount of color mixing is defined for each pixel, and the color mixing ratio is determined by measurement. Since it is necessary to maintain a sub-compensation for all the color mixing ratios of the entire pixel, as a suitable method, the present embodiment adopts the following P as a representative point, by performing a spline sPlne and a bezier between the representative points. The method of color mixing rate of linear interpolation. An illustration of a presumption of the color mixture ratio θ of the representative points of the map. In the image affix correction device 15, as shown in Fig. 14, the memory unit (5) stores an effective color mixture ratio 推 estimated by each of the areas divided into a matrix (mesh shape) in accordance with the imaging plane of the imaging element 12. The example of Fig. 14 is an example in which the matrix shape of Μ is divided, and the color mixture ratios θ 〇 Θΐ 。 9 are indicated. The color correction unit 151 calculates one of the pixel positions (coordinates) necessary for the color correction in the color ratio 保存 stored in the memory unit 153 by interpolation, and performs color mixture correction using the approximate color mixture ratio Θ. The color mixture correcting unit 15 1 performs color mixture correction according to the above formula 9. Fig. 5 is a schematic view showing a color mixture correction example (original principle) of the color mixture correcting unit 151 of the embodiment. In the example of Fig. 15, the memory unit 153 stores the color mixture correction coefficient representative point table TBL1 of each color. The color mixture correcting unit 151 reads the necessary color mixture ratio 由 from the memory unit 153 at the time of color mixture correction, and determines the color mixture ratio of each pixel by performing a spline, a Bézi curve, and a line interpolation between the representative points at a specific time point. 138321.doc 20· 201004370 After that, the color mixing correction unit (5) contends for the image signal of the clamp portion 14 of the 藉 by the product sum operator isn, and corrects the color mixing ratio of the I, 廿, I, and 疋The product is combined with the sub-hunting. The subtraction calculator 1512 is used to enter the ammonium 4D and the results of the difference and the subtraction method before the correction image is used to correct the color mixture. Fig. 16 is a view showing a shading correction (fourth) pattern of the embodiment. (10) Shadow correction example (In the example of Fig. 16, in the memory unit 154, the shading correction coefficient representative point table TBL2 is stored in each color of .. , T杈. The shading correction unit 152 is corrected by shading / ^ Ding Tian ° The hidden part 154 reads the necessary correction coefficient of the yin shirt, and determines the pixel of each point by the η 1~ clothing ··· quoting between the representative point, the Bezier curve and the linear interpolation at a specific time point. Then, the shading correction unit 152 performs the shading correction by integrating the image signal before the shading correction by the color correction unit (5) of the color correction unit (5) and the determined correction coefficient. The system diagram shows an example of the interpolation operation of the color mixture and the pixel shading correction in the present embodiment. In the interpolation operation, the representative point (the mesh unit) N is equally divided (the power of 2), as shown in FIG. Each mesh defines the weighting coefficients corresponding to the three representative points, and is obtained by [the weighting coefficient of the interpolation position of each representative point x]. Here, 'taken 0~1·〇, 1/N level=1 pixel level After that, the so-called Β spline interpolation is performed as follows. [Number 11] Interpolation position t0 Value = wl(t0). Represents the value of point 1 (Θ1) +w2(t0)_ represents the value of point 2 (Θ2) 138321.doc -21 201004370 + w3(t0)· represents the value of point 3 (Θ3) As described above, the image signal correcting device 15 of the present embodiment has the following features. The image positive device 15' is used to match the imaging surface of the imaging element 12: each region which is divided into a matrix (mesh shape) The effective color mixing (4), by interpolating the symplectic z τ color sheep and θ to the necessary pixel position (coordinate) θ, and performing the color mixture correction associated with Θ, like "correction device 15 pairs implementation The image signal after the color correction is corrected for shading. : Image: No. Correction device 15 'When mixed color correction and shadow correction, for example, shape), the complex of the mesh unit/object surface) is in a matrix form (Nets Π田之The so-called B-spline interpolation is performed on the weighting coefficients of the representative points. Therefore, according to the present embodiment, the following effects can be obtained. That is, the color mixture correction corresponding to the in-plane distribution can be realized. Not only do you consider the process of absorbing the stagnation, It is also considered. 'The correction of the degree can estimate the result of the correct signal, and the high-precision correction can be realized. Moreover, the method described in detail above can also be formed and executed by a computer such as a CPU. The procedure of 4 (2): This program can be composed of recording media such as semiconductor memory, magnetic disk, and optical monument (registered trademark), and Annon has the first [soft access to execute the above program." Fig. 1 is a block diagram showing an example of a configuration of an image pickup device of an image signal correction device according to an embodiment of the present invention; Fig. 2 is a view showing a configuration of an image pickup device of an image signal correction device according to an embodiment of the present invention; FIG. 3 is a circuit diagram showing a configuration example of a unit pixel of the imaging element of the embodiment; FIG. 4 is a schematic view showing a device configuration of the imaging element and an explanatory view of color mixing; Figure 6 is an image of the process of dissipating and absorbing the light; Figure 7 is a diagram showing the color mixing model between pixels; Figure 8 is a diagram showing the dissipation and absorption model of the light of each pixel. 9(A) and 9(B) are diagrams showing an example of an input/output signal of an image sensor (sensor); FIG. 10 is a flow chart showing signal processing of the image signal correction device of the embodiment; Fig. 1 is a view showing an example of the measurement system of the color mixture amount according to the embodiment; Fig. 13 is a view showing the response of each pixel of the monochromatic light R and the color mixture amount. Fig. 14 is an explanatory diagram showing an example of estimation of the color mixture ratio 0 of the dot; Fig. 15 is a schematic view showing a color mixture correction example (principle) of the color mixture correction portion of the embodiment; Fig. 16 is a shadow of the shadow correction portion of the embodiment. Schematic diagram of the correction example (principle); and 138321.doc • 23- 201004370 Fig. 1 7 shows an example of the interpolation operation of the color mixture and pixel shading correction in the present embodiment. [Description of main component symbols] 10 Imaging device 11 Lens system 12 Imaging device 13 A/D converter 14 Clamping unit 15 Image signal correction device 16 De-mosaic portion 17 Linear matrix portion 18 Gamma correction portion 19 Luminance color signal generation Portion 151 Color mixing correction unit 152 Pixel shading correction unit 153, 154 Memory unit 138321.doc 24-

Claims (1)

201004370 VII. Application for Patent Park: 1. An image signal correction device, which includes at least: a two-color positive portion, which accepts an image converted by a color image sensor to correct the color component contained in the image. And 5 杈 杈 ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' The color correction coefficient read by the part is obtained, and the color mixing ratio of the necessary pixel position is obtained, and the mixed color ratio is used to correct the color mixture component. 2. 3. 4. 5. If the image signal correction device of the requester is Let the above-mentioned color mixing correction system "capture the representative point of the number" to interpolate between the representative points to determine the color mixing ratio of each pixel. The image signal correction device of claim 1 or 2, wherein the color mixing correction P is a sum operation of the obtained color mixing ratio and the input image signal, and sub-processing the input image signal and the product sum operation result Perform color mixing correction. The image signal correcting device according to any one of claims 1 to 3, wherein the color mixture correction coefficient stored in said memory portion is measured and stored in advance as a color mixture ratio of another color pixel from one color pixel. The image signal correction device according to any one of claims 1 to 4, further comprising a shadow correction unit that receives an image signal corrected by the color mixture correction unit and performs pixel matching on the mixed color correction image signal Shadow correction. 138321.doc 201004370 6. The image signal correction device of claim 5, comprising: a second memory portion that stores (4) a pixel shading correction coefficient set in each of the plurality of regions corresponding to the imaging surface of the imaging element, and the above The shading correction unit acquires the correction coefficient of the necessary pixel position by interpolating and storing the pixel shading correction coefficient of the second pixel, and performs shading correction using the obtained correction coefficient. 7. The image signal correcting device of claim 6, wherein the shading correction and the correction coefficient obtained by the shading are integrated with the blended color correction image signal to perform shading correction. 8. An image pickup apparatus comprising: a color image pickup device that captures an image of a subject; and an image signal correction device that is corrected by an image signal photoelectrically converted by the color image pickup device; and the image signal correction device The method further includes: a color mixing correction portion that corrects a color mixture component included in an image signal photoelectrically converted by the color image sensor; and a memory portion that is stored in advance corresponding to the image pickup surface corresponding to the image pickup device, and each of the four (four) points The color mixing correction coefficient of the region; and the color mixing correction portion is added to the color mixing correction coefficient read by the upper trace memory portion by interpolation, thereby obtaining the color mixing ratio of the necessary pixel ,, and using the obtained color mixing ratio Correction of the color component of the feed. 9. The image pickup apparatus of the item 8 is configured to cause the color mixture correction unit to set a complex representative point of the 'interpolation representative' (four) to set the color mixture ratio of each pixel. The image pickup device of claim 8 or 9, wherein the color mixing correction unit performs a sum operation of the obtained color mixing ratio and the wheeled image signal, and performs a round-in image signal and the product-sum operation. The result of the subtractive processing is performed by the subtractive processing. The image-cracking of any one of claims 8 to 10, wherein the color mixture correction coefficient stored in the δ-resonation portion is used as another color pixel from a color pixel. The color mixing ratio is measured and stored in advance. 12. If request 34 to! ! The image pickup device according to any of the preceding claims, further comprising a shading correction unit that receives a color correction corrected by the color mixture correction unit: a signal 'subpixel shading correction of the mixed color correction image signal. 13. The imaging device of claim 12, comprising a first memory portion that stores a pixel shading correction coefficient that is set in advance in each of the plurality of regions corresponding to the imaging surface of the imaging element, and the shadow correction portion is borrowed The correction coefficient of the necessary pixel position is obtained by interpolating the pixel shading correction coefficient ' stored in the above-mentioned first memory portion, and the shading correction is performed by using the correction coefficient obtained far. The image processing device of the shading correction (4), wherein the shading correction unit performs an integration of the correction coefficient of the picking and the color correction image signal, and performs an image signal correction method, which includes: a step of reading a color mixture correction coefficient preset in each of the plurality of divided regions corresponding to the imaging surface of the color imaging element; and receiving an image signal photoelectrically converted by the imaging device, and reading the image by the memory portion by interpolation The color mixing correction coefficient is obtained, and the color mixing ratio of the pixel position of 138321.doc 201004370 is obtained; and the color mixing ratio obtained by the above-mentioned color mixing ratio is corrected. In the image processing, the image correction processing is performed on the computer, and the image correction processing includes the color correction coefficient previously set in the image plane of the corresponding color image sensor and the area of the number division. Processing; receiving an image signal photoelectrically converted by the imaging element, and interpolating the color mixing correction coefficient read by the memory unit, and processing the color mixing ratio of the pixel position of 3; and Color mixing rate for correction of color mixing components 〇138321.doc